Farida Fozdar, University of Western Australia
The Tower of Babel (Allen and Israel, 2021) is a compelling image when considering issues to do with translation and interpreting and the ethics of social research. Even when we speak the same language, we may not be ‘speaking the same language’, so to speak (excuse the triple metaphor). Talking past each other occurs in many ways but, in communicating the clear purpose and potential risks of one’s research, clarity is vital. Here, I outline a few issues from personal research experience, arguing that the communities themselves may be best placed to identify ethics issues and solutions to translation and interpreting dilemmas.
When working with those from a language different from that of the researchers, it may be the case that the idea of research is not well understood in the culture of origin. Therefore, explanations may need to start at a more fundamental level, with more time dedicated to ensuring participants understand the nature of research generally, as well as the specific project, expectations, rights and so on. Thus, rather than a shorter summary of the project than that offered to other participants, those using other languages may require more extensive explanation and information, provided in a cultural frame that makes sense to each group of participants. Often researchers and research ethics committees may be unaware of these cultural differences, leading to difficulties recruiting participants, and inadequate levels of understanding among those who do agree. On the other hand, research ethics committees can also take an overly paternalistic approach to ethnic minority populations, seeking to protect a presumed vulnerable set of participants, rather than taking a ‘layered’ approach that recognises internal diversity (e.g. second generation young people of refugee backgrounds with good English language skills and an understanding of the social research context) and agency (see Humpage et al, 2019). A one size fits all rule can’t be applied.
The issue of language and research ethics is complex, and ethics committees may need to defer to researchers about the best way to approach the issue in the circumstances of a particular project. Some years ago a seasoned colleague, seeking exemption from being required to provide a written information sheet and signed consent for some sensitive research in a nearby non-English speaking country, was asked by his institution’s committee to provide a written outline of what he would say to inform participants, with whom he had built a strong relationship of trust over decades, and who understood his research and its value, about the project. With tongue firmly in cheek, he wrote the information out and presented it to the Committee in the language in which he would be communicating with participants. It is unlikely the Committee would have had the expertise to understand the language, or paid for the document to be translated into English. Ultimately, the committee took it on faith that he would be conveying the information appropriately, something they could have done in the first place.
The gold standard in translation is backtranslation – to translate into the language of participants, and then have that translation translated back into English, to ensure that the meaning has not been distorted or lost in any way. In practice, however, it is difficult enough to secure funding to include one-way translation of ethics information, research instruments and responses, let alone backtranslation. Pragmatics often mean that researchers have to take the simpler route, managing with single translation and shorter information.
To get around prohibitive costs, researchers sometimes use bilingual workers employed as research assistants, rather than professional translators/interpreters. This may be appropriate, if they are being paid for this function, although the quality of translation and interpreting may not be ideal. However, sometimes family members or community volunteers are used. Such a practice can be exploitative, and also put these family or community members into awkward situations if issues of a sensitive nature are discussed. This is particularly the case where family members are minors, as is the case where children are recruited to interpret for parents. Interpreters should be neutral and passive, which may be difficult for family or community members; and family members may embellish, edit, or give their own version of events, rather than conveying the responses of the nominal participant. This practice should be avoided, and is often explicitly proscribed in medical research and practice. There are a range of complexities around the issue of who should translate/interpret. In research I was involved in some years ago, investigating mental health issues for migrants, bilingual assistants were used to interpret for members of minority ethnic communities. While information had been conveyed, in language, about the possible distress that discussing mental illness and its perceived causes might trigger (Tilbury, 2006), we found ourselves at a community gathering where we were reporting the results, watching women start to break down and cry as another woman recounted, in their shared language which we researchers could not understand, her experience with the Australian justice system of having her children taken away from her. Dealing with such a situation across languages (and of course cultures) was difficult, and not something that could have been anticipated by the researchers or the Committee. The bilingual assistant, a member of the community, knew the story, as did many of the women attending, and in this case it was probably useful that she was part of the community rather than an outsider provided by an official translation and interpreting service. Ultimately she was able to calm the situation, and the women approached us afterwards to emphasise how important the research was to convey their experiences and concerns to government.
Using official translation and interpreting services can also come with hidden risks. Sometimes interpreters will be sent from a rival ethnic community or one which may traditionally be in conflict with that of one’s participants, leaving them feeling threatened. Alternatively, if from the same group, the risk of breaches of confidentiality may be higher, and the likelihood of a frank and open response from the participant possibly lessened as a result of fear of such a breach, whether it actually occurs or not. For example, in small communities such as the 591 born in Eritrea or 233 from Laos or 310 Slovenians living in Western Australia (Office of Multicultural Interests, 2018), it may be unlikely for people not to know each other, even if they are officially employed in interpreting roles. In these cases, seeking interpreters from another region (e.g. interstate) would be appropriate, but even this may not guarantee confidentiality nor remove conflicts of interest.
Most important, however, is the recognition that not providing a translation and interpreting option can exclude certain groups from participation in research, limiting generalisability and further marginalising some populations.
What is the role of the research ethics committee in managing issues of translation and interpreting? A single hard line approach, for example, that insists on backtranslation by an official translator/interpreter from outside the geographical region, may not be practical, given the realities of access to interpreters, costs, written literacy levels, and so on. Often committees may have little expertise in these issues, and make demands of researchers that are unrealistic, or that challenge their ability to deal creatively and ethically with the situation as it arises. This can lead to the sort of tongue in cheek response by academics, prompted by frustration, that may not serve the interests of communities, committees, or academic peers. An approach that seeks to include the community itself in the design of the research is most appropriate to ensure ethical outcomes.
Allen, G. & Israel, M. (23 February 2021). The Tower of Babel and Human Research Ethics. Research Ethics Monthly. Retrieved from: https://ahrecs.com/the-tower-of-babel-and-human-research-ethics/
Humpage, L., Fozdar, F., Marlowe, J., & Hartley, L. (2019). Photovoice and refugee research: The case for a ‘layers’ versus ‘labels’ approach to vulnerability. Research Ethics, 15(3-4), 1-16. https://journals.sagepub.com/doi/10.1177/1747016119864353
Office of Multicultural Interests, Western Australia (2016). Country of birth and main languages spoken. 2016-census-language-data-country-of-birth-and-main-language-spoken.pdf (omi.wa.gov.au)
Tilbury, F. (2006). Ethical dilemmas: Principles and Practice in research with African refugees–a response to Bailes et al. Monash Bioethics Review, 23(4), 75-84. DOI: 10.1007/BF03351450
This post may be cited as:
Fozdar, F. (22 March 2021) Tongue in Cheek. Research Ethics Monthly. Retrieved from: https://ahrecs.com/tongue-in-cheek/